2

I have forest stand characteristic data from northern Sweden. We have interpreted multiple years of aerial photos and now have polygons outlining different forest stand ages over time (e.g. clear cuts, young forests, old forest, etc.). I'm interested in determining how the number of clear cuts has changed over time and how much area they have covered over time. But because of the way we mapped the forest stands, we can have 2 or more polygons that are called "clear cut" right next to each other. Thus, if I were to just count the number of polygons that were coded as "clear cut" this would over estimate the number of clear cuts. How can I get GIS (I'm a novice, but use both ArcGIS and QGIS) to count polygons that neighbor each other as just one polygon? aerial photo with selected polygons - all of which are designated as 'clear cut' but are in fact one big clear cut

  • So you want to count only each contiguous areas as a single clear cut? – Rex Sep 28 '17 at 13:33
5

Assuming you have attributes for the polygons for each type (clear cuts, young forest, etc), you can run the Dissolve tool in ArcGIS with this ‘Type’ column as the Dissolve_Field. This will aggregate all the features of each type. If you UNCHECK ‘Create Multipart features’ your result will be a single polygon for each contiguous feature of each type. This is also assuming each photo interpretation (time) is a different feature class. If not, and its all in one, you will need to also add the time (or whatever field differentiates the different photos) to the Dissolve_Field list.

  • 1
    Thanks Rex! I just checked this after not working on the project for awhile and this is perfect! – EMH Nov 7 '17 at 14:12
2

In QGIS, the Dissolve tool (in the tob bar, Vector > Geoprocessing Tools > Dissolve) is used to join adjacent polygons.

2

This seems better suited for a raster data type. Convert each polygon dataset into a raster where, clear-cut is coded as 1 and everything else 0. Then, just sum your rasters.

The specified resolution of the raster(s) would depend on the mapping base scale of your polygons (eg., 1:50,000 would support ~25m resolution). You can approximate this using: [Map base scale = raster resolution (meters) * 2 * 1000] so, a 30m raster resolution would be supported at 1:60,000.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.